1. Technical Field
The inventions disclosed herein are directed to systems, devices and methods for establishing an interface through the thickness of the skull for purposes such as delivering some form of neuromodulation (e.g., electrical or optical stimulation, pharmaceutical stimulation or thermal (e.g., cooling or delivering ultrasound to the brain)) to targeted structures in the brain in a controlled manner to modulate neural activity, and detecting signals generated by neurons in targeted structures in the brain.
2. Background
Much research and clinical development activity is ongoing in the area of using various forms of neuromodulation to affect the brain (e.g., to diagnose or treat a neurological disorder). There is also continuing interest in improving the quality or fidelity with which signals can be sensed or measured from the brain, especially in electroencephalography but also with respect to measurements associated with things such as impedance plethysmography, tomography, and optical imaging.
Electrical Stimulation
Neuronal activity can be measured as electrical signals. This activity also can be modulated (e.g., to inhibit undesired activity by blocking the action potentials that allow the neurons to “fire”, to increase or decrease the excitability of a group of neurons, or to cause neurons to fire) by inducing an electric field in neural tissue, or stated another way, in the vicinity of a group of neurons.
One way of inducing an electric field is by conducting electricity to the neural tissue through an electrode-to-tissue interface (ETI). Implantable and partially implantable systems are known which can deliver neuromodulation in this manner. For example, U.S. Pat. No. 6,016,449 to FISCHELL et al., issued Jan. 18, 2000 for a “System for Treatment of Neurological Disorders” describes an implantable neurostimulation system which, through electrodes implanted on the surface of or in the brain, detects signals (referred to as electrocortical signals or “ECoG”s because they are measured directly at the brain as opposed to through the skull, as is the case with a conventional electroencephalogram). The system can be configured so that, when the neurostimulator detects certain types of activity in the ECoGs, e.g., activity that is believed to be associated with a seizure or to be a precursor of a seizure, it will deliver electrical stimulation to targeted areas of the brain in the form of various types of electrical waveforms, with the intention of eliminating seizure activity and/or reducing the severity of the seizures.
The types of waveforms that can be delivered through an electrode-to-tissue interface are limited inasmuch as the charge density per phase has to be low enough to be considered safe and charge balancing must occur. More specifically, in a conventional electrode, current is carried by movement of electrons within the electrode, typically a metallic substance. However in an aqueous, non-metallic environment such as the human body, current is created largely by the movement of ions (charged particles) within the environment. In order for electrical charge delivered by an electrode to pass into and affect the surrounding tissue, the electric current flowing through the electrode must be converted into ion movement in the tissue.
This conversion can happen in two ways, by virtue of capacitance or electrochemical reactions.
More specifically, an electrode interface, such as an electrode-to-tissue interface, is capacitive; that is, it can store a small amount of electrical charge without any actual transfer of charge from electrode to tissue. Consider two pipes attached end-to-end with a rubber membrane separating them. A small amount of flow in one pipe can balloon out the membrane, and, as long as the amount of flow is not great enough to burst the membrane, the net flow of current is transferred to the second pipe. If the flow is then repeatedly reversed (the analogy here being to alternating current), the system appears as if it were one single pipe with no barrier. This occurs, electrically, when small amounts of charge are delivered in a biphasic pulse; the leading phase stores charge in one direction, and the trailing phase removes charge to restore the balance.
If the electrical charge to be passed exceeds the capacitive limit of the electrode-to-tissue interface, then the only remaining way to transfer charge is by electrochemical reactions occurring at the electrode-to-tissue interface. The precise nature of the reactions that occur depend on the voltage across the ETI, but the reactions are almost always undesirable because they can result in, for example, hydrogen ions, hydrogen gas, hydroxide ions, oxygen gas, and other possibly toxic substances being introduced into the tissue, and denaturation of proteins already present in the tissue. The reactions can also result in erosion of the electrode and distribution of the electrode material into the surrounding tissue. Current passed in this way is often referred to as Faradaic current. For the sake of completeness, it is noted that in practice, a small amount of reaction product can be absorbed by a reversed electrochemical reaction on the trailing phase of a biphasic stimulus. This is known as pseudocapacitance. The actual safe charge per phase of a stimulation system (i.e., the amount of charge per phase that does not result in undesirable electrochemical changes in the tissue over time) thus is governed by the sum of the actual capacitance and the pseudocapacitance.
In view of the foregoing, a goal associated with use of electrical neurostimulation systems using implanted electrodes is to keep the charge passed per phase within the capacitive limit of the ETI. The magnitude of this limit is key to safety of electrical neurostimulation, and has been characterized for some materials from which electrodes are commonly fabricated. Platinum, for example, yields a theoretical charge storage capacity of 200 μC/cm2 (micro coulombs per square centimeter) and a practical charge storage capacity of 50 μC/cm2. Oxide materials such as iridium oxide may reversibly store more than 1000 μC/cm2. These charge densities are more than sufficient for pulsatile or high frequency stimulation in most cases. By comparison, a typical deep brain stimulus of 3 mA for a 90 μS pulse width on a 5.7 mm2 electrode passes 5 μC/cm2. On the other hand, low frequency, non-pulsatile electrical stimulation is constrained in most cases by these limits. For instance, a 1 Hz sinusoid delivered at 1 mA peak-to-peak on a 5.7 mm2 electrode passes 2800 μC/cm2 per phase. (GRILL, W. M., “Safety Considerations for Deep Brain Stimulation: Review and Analysis,” Expert Rev. Med. Devices (2005), 2(4): 409-420 and MERRILL, D. R., et al., “Electrical Stimulation of Excitable Tissue: Design of Efficacious and Safe Protocols,” J. Neurosci. Meth. (2005), 141: 171-198.)
Accordingly, the waveforms used in electrical stimulation delivered through an electrode-to-tissue interface are those which can both maintain charge balancing and either avoid or reverse any electrochemical reactions at the ETI as they begin to occur. Stimulation using waveforms that satisfy these criteria will be referred to herein as “pulsatile stimulation” or “AC stimulation.” Examples of these waveforms are biphasic pulsatile waveforms (as are commonly used for deep brain and cortical neurostimulation) (see MERRILL et al., “Electrical Stimulation of Excitable Tissue: Design of Efficacious and Safe Protocols,” J. Neurosci. Meth. (2005) 141: 171-198), and sinusoidal or near-sinusoidal waveforms at high frequencies such as 100 Hz and above.
There are also neuromodulation techniques which rely only upon an external stimulation source and which are believed to modulate neural activity by inducing a current in neural tissue. One of these techniques is Transcranial Electrical Stimulation or “TES.” TES involves applying electrodes to the scalp which, when provided with an electrical signal, result in some current flow in the brain which in turn has the effect of modulating the activity of groups of neurons. TES is usually not a preferred approach to treating a disorder or other condition of a patient, because most of the current from the stimulation flows through the scalp, from electrode to electrode, rather than into the brain, and this current flow causes pain and discomfort, due to stimulation of nerves in the scalp, and contraction of the scalp muscles. It has been used as a form of electroconvulsive therapy (ECT) with the patient under anesthesia, to treat depression.
Transcranial Direct Current Stimulation (tDCS) is another technique to modulate the electrical activity of neurons. In this technique, weak electrical currents (on the order of 0.1 to two milliamps) are applied through electrodes placed externally on the scalp, with conduction to the scalp facilitated by a saline-saturated sponge or a layer of conductive gel. The currents, and the resulting static DC fields, are believed to alter the firing rates of neurons. tDCS is being investigated for use in treatment of several conditions; for example, major depression. For example, in one reported double-blind study, anodal tDCS was applied to the left dorso-lateral prefrontal cortex and was observed to improve mood in 40 patients when compared to both anodal tDCS applied to the occipital cortex (believed to be unrelated to depression) and sham stimulation. (See BOGGIO, P. S., et al., “A Randomized, Double-Blind Clinical Trial on the Efficacy of Cortical Direct Current Stimulation for the Treatment of Major Depression,” Int. J. Neuropsychopharmacol (2007) 11, 1-6. Another study has reported improved go-no-go task performance in depressed patients using a similar protocol. (See BOGGIO, P. S., et al., “Go-No-Go Task Performance Improvement after Anodal Transcranial DC Stimulation of the Left DorsoLateral Prefrontal Cortex in Major Depression,” J. Affect Disord. (2006) 101(1-3): 91-8.
tDCS also has been used experimentally to treat a variety of neurological disorders, as well as in experiments designed to study and enhance cognitive function in normal human subjects. Most studies have concluded that tDCS has a mild neuromodulatory effect, often of clinical value and often lasting beyond the immediate stimulation period. A scientific review of experimental, human clinical use of tDCS is provided in FREGNI, F., et al., “Technology Insight: Noninvasive Brain Stimulation in Neurology—Perspectives on the Therapeutic Potential of rTMS and tDCS,” Nat. Clin. Pract. Neurol. (2007) 3(7): 383-93). There are some articles in the popular press on the subject as well, such as TRIVEDI, B., “Electrify Your Mind—Literally,” New Scientist, 15 Apr. 2006, and KENNEDY, P., “Can A Jolt From A Nine-Volt Battery Make You Smarter? Happier? Medical Researchers Revive A Discarded Technology and Set the Stage for the ‘Brain Pod’,” The Phoenix, 7 Feb. 2007.
For example, stroke rehabilitation using tDCS, particularly rehabilitation for strokes that caused some type of motor deficit, has been studied by several groups. Anodal tDCS, applied to the area of an ischemic lesion, improved standard measures of motor function in a sham-controlled group of six patients with mild motor deficit (as disclosed in HUMMEL, F., et al., “Effects Of Non-Invasive Cortical Stimulation On Skilled Motor Function In Chronic Stroke,” Brain (2005) 128:490-00) and in a group of eleven patients with severe motor deficit (as disclosed in HUMMEL, F. et al., “Effects of Brain Polarization On Reaction Times And Pinch Force In Chronic Stroke,” BMC Neuroscience (2006) 7:73.) In this and other applications, anodal tDCS is believed to be excitatory, increasing cortical excitability and enhancing neural plasticity in the stimulated region. The effect is believed to last somewhat beyond the actual stimulation session.
Further, application of cathodal tDCS to the area contralateral to an ischemic lesion, in addition to anodic tDCS to the lesion area, has been observed to similarly improve motor function in six patients with mild to moderate motor deficit (as disclosed in FREGNI, F., et al., “Transcranial Direct Current Stimulation Of The Unaffected Hemisphere In Stroke Patients,” Neuroreport (2005) 16: 1551-1555.) In this and other applications, cathodal tDCS is believed to be inhibitory, decreasing cortical excitability and in particular decreasing output of the stimulated region.
Cathodal tDCS, applied over an epileptic cortex, has been shown in at least one report to reduce the number of epileptiform discharges observed within 30 days after stimulation (as disclosed in FREGNI, F., et al., “A Controlled Clinical Trial Of Cathodal DC Polarization In Patients With Refractory Epilepsy,” Epilepsia (2006) 47(2): 335-342). A trend toward reduced seizure frequency, i.e., not reaching the level of p=0.05 significance, was also observed after cathodal tDCS. It was noted in this study that anodal tDCS, applied over the contralateral, non-epileptic cortex, did not cause increased epileptiform discharges. A similar treatment currently is the focus of a trial sponsored by the National Institute of Neurological Disorders and Stroke for 56 patients (see “Anticonvulsive Effects of Transcranial DC Stimulation in Pharmacoresistant Focal Epilepsy,” NIH Protocol No. 06-N-0187 (2006).)
Neurostimulation using pulsatile waveforms applied to the motor cortex has been used for treating chronic pain, especially for pain of neuropathic or central origin. Using tDCS to treat such pain has also been reported. In one study of 17 patients (as disclosed in FREGNI, F., et al., “A Sham-Controlled, Phase II Trial Of Transcranial Direct Current Stimulation For The Treatment Of Central Pain In Traumatic Spinal Cord Injury,” Pain (2006) 122: 197-209), anodal tDCS over the primary motor cortex was shown to significantly reduce pain due to fibromyalgia when compared to both sham stimulation and anodal stimulation of the dorso-lateral prefrontal cortex (DLPFC, an area of cortex which is thought to be unrelated to the condition of central pain).
Some are investigating using tDCS for treatment of the movement disorder Parkinson's disease. One report suggests beneficial effects on motor-task scores and motor-evoked potentials in 17 Parkinsonian patients (FREGNI, et. al., “Noninvasive Cortical Stimulation with Transcranial Direct Current Stimulation in Parkinson's Disease,” Mov. Disord. (2006) 21: 1693-1702.
Still another promising area of tDCS research involves cognitive enhancement in normal human subjects. tDCS administered during slow-wave sleep has been observed to increase retention of memorized word pairs significantly, in comparison with both sham stimulation and tDCS administered in those who are awake. (See MARSHALL, L., et al., “Transcranial Direct Current Stimulation during Sleep Improves Declarative Memory,” J. Neurosci. (2004) 24 (44): 9985-9992, as corrected in J. Neurosci. 25(2).)
Fregni et al. also observed enhanced performance with a working memory task, in 15 subjects, with anodic tDCS applied over the left dorso-lateral prefrontal cortex. (FREGNI, F., et al., “Anodal Transcranial Direct Current Stimulation Of Prefrontal Cortex Enhances Working Memory,” Exp. Brain Res. (2005) 166(1): 23-30.) This enhanced performance was contrasted to cathodic stimulation of the left DLPFC, which had no effect, and anodic stimulation of the primary motor cortex, which also had no effect and which is believed to be an area of cortex irrelevant to working memory. In another study, Marshall et al. identified significant slowing of reaction time in 12 subjects with bilateral frontal tDCS, during a working memory task (MARSHALL et al., “Bifrontal Transcranial Direct Current Stimulation Slows Reaction Time In A Working Memory Task,” BMC Neuroscience (2005) 6:23.)
Administration of anodal tDCS over left prefrontal cortex has also been shown to significantly increase verbal fluency in contrast with cathodal tDCS, which resulted in a mild decrease in fluency. (IYER, M. B., et al., “Safety and Cognitive Effect of Frontal DC Brain Polarization in Healthy Individuals,” Neurology (2005) 64(5): 872-5.)
In addition, one study suggests that alcohol craving can be decreased using anodic-left/cathodic-right and anodic-right/cathodic-left tDCS of the dorso-lateral prefrontal cortex. (BOGGIO, P. S., “Prefrontal Cortex Modulation Using Transcranial DC Stimulation Reduces Alcohol Craving: A Double Blind, Sham-Controlled Study,” Drug Alcohol Depend, 17 Jul. 2007.) The effect was demonstrated in 13 subjects to be significant in comparison to sham stimulation, regardless of tDCS polarity.
Some of the difficulties facing researchers investigating various applications of tDCS relate to the ability to focus the stimulation on target areas of the brain and the ability to accurately or repeatedly locate the scalp electrodes to provide the desired stimulation.
Modeling of current and electrical field distribution in tDCS shows that electrical fields sufficient for neuromodulation are widely distributed throughout the brain. (See LU, M, et al., “Comparison of Maximum Induced Current and Electric Field from Transcranial Direct Current and Magnetic Stimulation of A Human Head Model,” PIERS Online 3(2) (2007) 179-183.)
This is significant, since most applications or potential applications of tDCS will require stimulation of a defined cortical structure, such as the primary motor cortex. Even those applications which involve providing diffuse stimulation of a larger structure, such as the dorsolateral prefrontal cortex, will likely target that structure only, such that stimulation of nearby structures would not be optimum.
Another issue in tDCS may be unfocused and/or undesired stimulation due to the reference electrode. While such stimulation may be mitigated somewhat by placing the reference electrode away from the patient's head, such placement may raise other issues. For example, placing the reference electrode elsewhere may result in unintended neuromodulation of the brain stem, due to the diffuse nature of the current flow. (See NITSCHE, M. A., et al., “Modulation of Cortical Excitability by Weak Direct Current Stimulation—Technical, Safety and Functional Aspects,” Supp. Clin. Neurophysiol. (2003) 56: 255-76.)
There have been several attempts to address the focality issue with tDCS. It has been shown that smaller stimulating electrodes and larger reference electrodes contribute to focal stimulation. (See NITSCHE, M. A., et al., “Shaping The Effects Of Transcranial Direct Current Stimulation Of The Human Motor Cortex,” J. Neurophysiol. (2007) 97:3109-3117.) Concentric ring electrodes have also been proposed, and used in an animal model, to provide more focused transcranial DC stimulation and to reduce reference electrode effects. (See BESIO, W. G., et al., “Effects Of Noninvasive Transcutaneous Electrical Stimulation Via Concentric Ring Electrodes On Pilocarpine-Induced Status Epilepticus In Rats,” Epilepsia, 25 Jul. 2007.) Accurate mapping of electrical properties of the head, and finite element modeling of tDCS current flow, has also been proposed as a way to increase focality of tDCS. (U.S. Patent Application Publication No. 2007/0043268, “Guided Electrical Transcranial Stimulation (GETS) Technique,” to RUSSELL, Feb. 22, 2007.) However, all of these techniques are still fundamentally limited by current preferentially flowing through the scalp, and blurring of the intracranial neuromodulatory field due to high skull resistivity. This is analogous to the situation in EEG; resolution is incrementally improved by using more electrodes but a fundamental limit is soon reached, with diminishing returns after about 2.5 cm inter-electrode spacing, due to blurring of the signal by intervening tissue. (See SRINIVASAN, R., “Methods To Improve The Spatial Resolution Of EEG,” Int. J. Bioelectromagnetism (1999) 1(1):102-111.)
Other techniques for applying electrical stimulation to the brain are under investigation that use waveforms (as opposed to direct current) which do not meet the definition of “pulsatile” or “AC” set forth above, i.e., the waveforms are not suitable for maintaining charge balance and for minimizing undesirable electrochemical reactions at the electrode-to-tissue interface. Stimulation using these waveforms will be referred to in this disclosure as “non-pulsatile stimulation” or “near-DC stimulation.” Examples of these waveforms are large amplitude or slowly varying oscillatory waveforms and low frequency sinusoidal waveforms. The nature of these waveforms is such that they exceed the limits of charge density per phase that are deemed safe at the electrode-to-tissue interface or they do not permit charge balancing to be maintained when the waveforms are delivered. Low frequency sinusoidal stimulation has shown some efficacy in animal models of epilepsy. (See GOODMAN, J. H., et al., “Low-Frequency Sine Wave Stimulation Decreases Seizure Frequency In Amygdala-Kindled Rats,” Epilepsia (2002) 43 (supp7): 10, and GOODMAN, J. H., et al., “Preemptive Low-Frequency Stimulation Decreases The Incidence Of Amygdala-Kindled Seizures,” Epilepsia (2005) 46(1): 1-7.)
In summary, then, the sources for electrical stimulation discussed above can be conveniently (for the purposes of this disclosure) grouped into these categories: (1) pulsatile or AC stimulation; (2) DC stimulation; and (3) non-pulsatile and near-DC stimulation. Applying electrical stimulation to modulate neural activity through an electrode-to-tissue interface typically requires invasive surgery to implant the electrodes, e.g., deep in the brain, on the cortex (cortical electrodes), or on the dura (epidural electrodes). The type of stimulation that can be delivered through the electrodes is limited, as a practical matter, to pulsatile or AC stimulation, because the waveforms used have an acceptable charge-density-per-phase and maintain charge balancing when delivered. Non-pulsatile or near-DC stimulation and direct current stimulation should not be applied through an implanted electrode-to-tissue interface because of unacceptable charge densities and the inability to maintain charge balancing during delivery. Without the implanted electrode-to-tissue interface, however, focusing the stimulation where it is desired to modulate neural activity is difficult, since the resistance of the skull tends to diffuse the electrical fields so that they are widely distributed throughout the brain. In addition, in tDCS, locating the scalp electrodes inaccurately can lead to errors in delivery of the stimulation, or in interpreting the results.
Magnetic Stimulation
Another technique for neuromodulation that is being explored is referred to as Transcranial Magnetic Stimulation or “TMS.” TMS is thought to induce eddy currents in the surface of the brain that stimulate a group of neurons. In this technique, the coil of a magnet is held against the head and energized by rapidly discharging a capacitor, which creates a rapidly changing current in the coil windings. This rapidly changing current sets up a magnetic field at a right angle to the plane of the coil. The magnetic field goes through the skin and skull to the brain and induces a current tangential to the skull. This current influences the electrical activity of the neurons. TMS can be applied on a single-pulse or paired-pulse basis, or repetitively (rTMS). TMS is not associated with the often high level of discomfort that accompanies TES. However, TMS is not favored in surgical environments because of the difficulties presented by having multiple metal objects in the environment. In addition, when used in any environment, TMS equipment is typically bulky to manipulate and consumes a lot of power. Also, the stimulation parameters in TMS tend to be less consistent than those that can be achieved with other types of electrical stimulation. TMS is under investigation for treatment of migraine headaches, amblyopia, and depression, among other neurological disorders and conditions.
Neuromodulation Using Lontophoresis
Lontophoresis refers to the techniques of moving an ionically-charged substance into and through tissue by electromotive force. The basic technique is well known, and has been used for delivering such biologically active agents (also known as bioactive agents) as anti-inflammatory medications, and topical anesthetics. Bioactive agents intended to affect neural tissue also can be delivered via lontophoresis. These agents may include but are not limited to glutamate, acetylcholine, valproate, aspartate, and gamma amino butyrate. Reverse lontophoresis (i.e., extraction of substances, usually for measurement) also is a well known technique in some applications as glucose monitoring. (See, e.g., RHEE, S. Y., et al., “Clinical Experience of an lontophoresis Based Glucose Monitoring System,” J. Korean Med. Sci. (2007) 22:70-3.)
Jacobsen, et al. describe early improvements for safety and comfort of lontophoresis and applications such as transdermal delivery of pilocarpine for diagnosis of cystic fibrosis, and transdermal delivery of anesthetic substances. U.S. Pat. No. 4,141,359 to JACOBSEN et al. for “Epidermal lontophoresis Device,” issued Feb. 27, 1979.
Using waveforms other than DC waveforms in iontophoresis are also known. For example, Liss et al., describes a three-component modulated waveform, reviews other iontophoretic waveforms, and presents results for iontophoretic diffusion of the bioactive substances adrenocorticotropic hormone (ACTH), cortisol, beta endorphin, and serotonin. U.S. Pat. No. 5,421,817 to LISS et al. for “Non-Intrusive Analgesic NeuroAugmentive and Iontophoretic Delivery Apparatus and Management System,” issued Jun. 6, 1995.
Lontophoresis also has been recognized as a promising avenue for delivery of substances into the brain. Lerner has proposed iontophoretic administration of pharmaceuticals into the brain tissue via transnasal or transocular paths. U.S. Pat. No. 6,410,046 to LERNER for “Administering Pharmaceuticals to the Mammalian Central Nervous System,” issued Jun. 25, 2002.
Lemer further has disclosed iontophoretic administration of bioactive substances to the central nervous system, using a source of bioactive substance that may be implanted at the brain surface. U.S. Pat. No. 7,033,598 to LERNER for “Methods and Apparatus for Enhanced and Controlled Delivery of a Biologically Active Agent into the Central Nervous System of a Mammal” issued Apr. 25, 2006. In addition, Abreu has disclosed iontophoretic delivery of substances via a naturally-occurring physiologic “brain-temperature tunnel” or “BTT.” U.S. Pat. No. 7,187,960 to ABREU for “Apparatus and Method for Measuring Biologic Parameters,” issued Mar. 6, 2007.
Stimulation Using Light
Techniques using light to modulate the activity of genetically modified neural tissue are well known. (See, e.g., DEISSEROTH, K., et al., “Next-Generation Optical Technologies for Illuminating Genetically Targeted Brain Circuits,” J. Neurosci. (2006) 26(41): 10380-10386.)
Detecting Brain Activity
EEG
The electroencephalograph, or EEG, is a measurement of scalp potentials resulting from the summed electrical contributions of many neurons. Poor spatial resolution of scalp EEG, due to spatial “blurring” of the signal by the relatively nonconductive skull, is a well known and well understood issue; maximal scalp EEG resolution is on the order of several centimeters, and decreasing inter-electrode spacing past approximately one centimeter yields virtually no improvement, since the signals are already “blurred” by the time they reach the scalp.
Mathematical models such as the spline-Laplacian and dura imaging have been described for preferentially extracting high-spatial-frequency information from the scalp EEG. (See, e.g., NUNEZ, P. L., et al., “A Theoretical and Experimental Study of High Resolution EEG Based on Surface Laplacian and Cortical Imaging,” Electroencephalogr. Clin. Neurophysiol. (1994) 90(1): 40-57; NUNEZ, P. L., et al., “Comparison Of High Resolution EEG Methods Having Different Theoretical Bases,” Brain Topogr. (1993) 5(4): 361-4.) These methods, however, still rely on a signal in which high-resolution spatial information is largely lost, and due to fundamental mathematical issues, provide no way of unambiguously reconstructing the unblurred signal.
By placing electrodes directly on the cortex or dura, one can measure electrical signals without experiencing the blurring caused by the skull and tissue that otherwise intervenes between the dura and the external scalp surface. This electrocorticograph, or ECoG, contains significantly more information at fine spatial scales than can be obtained with scalp-recorded signals. Obtaining ECoGs, however, requires invasive surgery to place the electrodes on the cortex or dura. This method of acquiring EEG is usually limited either to acute use or with a chronically implanted device. In acute use, wires are typically run from the electrodes on the cortex through the skin to an external amplifier (e.g., for mapping epileptic foci over several days or weeks). This technique is associated with a risk that the wires or electrodes will become dislodged and that the exposed area may become infected.
If a chronically implanted ECoG detector is used (such as that disclosed in U.S. Pat. No. 6,016,449 to FISCHELL et al., issued Jan. 18, 2000 for a “System for Treatment of Neurological Disorders”), the risk of dislodgment and infection is lessened. However, to implant the device and the electrodes and associated leads is invasive and expensive. Power and other design constraints may limit the extent to which implanted devices can process ECoGs without external equipment.
Impedance Plethysmography and Tomography
Electrical impedance plethysmography is a well-known method for estimating the volume of an anatomical space by measuring electrical impedance at various frequencies. It may be used to measure volumetric or density changes in neural and vascular tissue, such as changes in perfusion, that are associated or thought to be associated with changes in neural activity.
A map of brain plethysmographic changes may be reconstructed from multi-channel scalp impedance measurements, a technique which is called Electrical Impedance Tomography or “EIT”. (See, e.g., CLAY, M. T., et al., “Weighted Regularization in Electrical Impedance Tomography with Applications to Acute Cerebral Stroke,” IEEE Trans. Biomed. Eng. (2002) 21(6): 629-637.) This plethysmographic signal, as measured on the scalp, is blurred in the same way that EEG signals are spatially blurred. Thus, reconstruction of even a crude tomographic image is both mathematically complex and not highly accurate.
Optical Imaging and Tomography
Optical methods for measuring brain activity such as cerebral perfusion and cerebral hemodynamics are well known. Optical sensing currently has been implemented for such things as direct optical recording of intrinsic reflectance signals (ORIS) from the surface of the brain, and transcranial optical tomography, which attempts to mathematically reverse scattering of light to skull and scalp tissue, in order to reconstruct a crude image of brain hemodynamics. (See, e.g., SUH, M., et al., “Blood volume and hemoglobin oxygenation response following electrical stimulation of human cortex,” Neurolmage (2006) 31:66-75, and HEBDEN, J. C., et al, “Three-dimensional optical tomography of the premature infant brain,” Phys. Med. Biol. (2002) 47:4155-4166.)
Transferring Energy to and from the Brain
Investigations have suggested that removing energy from the brain may have application in eliminating or reducing the severity of neurological disorders. For example, heat transfer has shown some promise as a technique for neuromodulation. More specifically, heat transfer away from a region of brain tissue (i.e., cooling) is known to reversibly deactivate neural tissue (i.e., the deactivated tissue can be reactivated after the cooling source is withdrawn), and has been shown to suppress spontaneous epileptiform activity in humans. This phenomenon is believed to provide a potential treatment for focal epilepsy. (See, e.g., KARKAR, et al., “Focal Cooling Suppresses Spontaneous Epileptiform Activity without Changing the Cortical Motor Threshold,” Epilepsia (2002) 43(8): 932-935.) However, engineering of a practical implantable cooling device has proven non-trivial. (See, e.g., ROTHMAN, et al., “Local Cooling: A Therapy for Intractable NeoCortical Epilepsy,” Epilepsy Curr. (2003) 5(5): 153-56.) Thermoelectric Peltier devices appear to offer promise, but are known to be relatively inefficient. Provision of adequate power, and safe disposal of the resulting heat, are practical design constraints that militate against an implantable cooling system.
Transcranial heat transfer has also been analyzed. (See, e.g., SUKSTANSKII, A. L., et al., “An Analytical Model of Temperature Regulation in Human Head,” J. Themr. Biol. (2004) 29:583-587.) Surface head cooling can be used during bypass surgery to induce hypothermia, improving low oxygen survivability and affording more time in which to accomplish the bypass procedures. Similarly, the surface of the head can be cooled to reduce inflammation or for other purposes. However, the cooling effect on the brain is usually nominal using this technique, and it is hard to focus, in any event, especially in the presence of normal blood flow.
High intensity focused ultrasound (“HIFU”) can ablate tissue deep in the body. It has been used to create lesions in the heart to treat atrial fibrillation and to ablate fibroid tumors. Using HIFU to treat the brain may be desirable insofar it is less invasive than open brain surgery, which may be complicated by neurological deficits, among other things. The skull, however, is a difficult barrier through which to deliver ultrasound energy, because the skull bone has a strong defocusing effect on the externally applied energy. Sophisticated techniques are being investigated to help overcome the ultrasound-scattering effect of bone, such as time-reversal mirrors (See, e.g., TANTER, M., et al., “Time Reversal for Ultrasonic Transcranial Surgery And Echographic Imaging,” Abstract, Acoustical Society of America J. (2005) Vol. 118; Issue 3, p. 1941), although skull heating during delivery of the ultrasound still presents an obstacle to this type of treatment.
Skull/Brain Interfaces
There have been some methods and devices disclosed for providing an interface through the skull to the brain as an alternative to, on the one hand, external stimulation sources or sensing electrodes and, on the other hand, implanted electrodes with associated implanted or partially implanted equipment.
For example, Lowry et al. have proposed positioning adjustable length intracranial electrodes through the thickness of the skull under local anesthesia, wherein a distal surface or extension of the electrode is adapted to electrically contact a surface of the brain, such as the dura mater, the cerebral cortex, or a deep brain structure. The electrode is then electrically connected to a pulse generator to apply electrical neurostimulation. U.S. Patent Application Publication No. 2004/0102828, published May 27, 2004 to LOWRY et al. for “Methods and Systems Employing Intracranial Electrodes for Neurostimulation and/or Electroencephalography.” In one embodiment, Lowry et al. discloses using an electrically conductive elastomer in an intracranial electrode, e.g., a polymeric material filled with a suitable quantity of a conductive metal powder. Lowry et al. also does not disclose using anything other than pulsatile stimulation from a pulse generator to generate electrical neurostimulation of structures in the brain. (See, e.g., U.S. Patent Application Publication No. 2004/0102828 [0077].)
Lowry et al. have also proposed an intracranial electrode having a head and a shaft such that a proximal portion of the head is flush with the outer layer of the skull. U.S. Patent Application Publication No. 2005/0075680, published Apr. 7, 2005 to LOWRY et al. for “Methods and Systems for Intracranial Neurostimulation and/or Sensing”. Lowry et al. also disclose an intracranial electrode with an “electrical energy transfer mechanism” or “ETM” that is placed externally adjacent a patient's scalp to couple electrical energy from a pulse generator to an intracranial electrode having a core using an electrically conductive material in conjunction with a conductive gel layer in an intracranial electrode system. (See, e.g., U.S. Patent Application Publication No. 2005/0075680 [0138]-[0141], FIGS. 34A & B and FIG. 39.)
Fowler et al. have proposed a method using an electrode implanted in a patient's skull to transfer stimulation signals (e.g., from a pulse generator) through the scalp to a target neural population. U.S. Patent Application Publication No. 2006/0106430, published May 18, 2006 to FOWLER et al. for “Electrode Configurations for Reducing Invasiveness and/or Enhancing Neural Stimulation Efficacy, And Associated Methods.”
The Lowry et al. and Fowler et al. references disclose only skull/brain interfaces through which signals are delivered or sensed using metal as a conductor. In addition, the only form of stimulation deliverable by any of the devices disclosed in the Lowry et al. references is pulsatile electrical stimulation.